Charter, changes in prefixes, and documentation

John C Klensin klensin at
Wed Mar 26 01:18:51 CET 2008


In the hope of moving forward --i.e., of bringing the charter
discussion to a conclusion and getting on to the actual work, a
few comments and then what I hope will be a generally-acceptable

I think we have gotten confused about the difference between
"things the WG cannot discuss or document without rechartering"
and "things the WG can't agree to do without rechartering and
getting more general IETF approval in the process".   At least I
have.  In particular, I think the discussion about why we don't
want to change prefixes, what the consequences of such changes
might be, and what incompatibilities we are willing to introduce
and live with without making such a change is important and
should be captured in one of the documents (presumably
"Rationale" if we don't change structure).  I think that should
be done both for the information of people making the transition
and to reduce the odds of some future discussion having to
repeat this discussion.  

At the same time, it seems clear that the WG should not be
making prefix changes without a much broader review, going in,
of the implications of doing that.


(1) As far as the charter is concerned, let's be explicit about
this.  Rather than scattering language and restrictions
throughout the charter, let's make an explicit subsection that
says something like 

	The WG will stop, close, and recommend that a new
	charter be generated if it concludes that any of the
	following are necessary to meet its goals:
		(i) A change to the "punycode" algorithm or the ACE
		approach to encoding names in the DNS
		(ii) A change to the ACE prefix from "xn--"
		(iii) A change to the basic approach taken in the design
		team documents.

I think that is the entire list, but, if others have other
items, let's get them on the list.

(2) We will work together to improve the description of the
prefix change issue in "Rationale".  I've now got a few ideas
about how to improve that text, but will need more input on this.

(3) I will work with Simon and others who are interested to be
sure that every incompatible change is discussed carefully and
that the discussion includes at least some recommendations as to
what to do about each one.   That material can go into Rationale
or, if the WG prefers, we can pull it out into another (fifth?)
document.   I don't think we need to decide on that before
chartering, but the charter text should probably indicate that
we will be explicit about any incompatible changes.

(4)  While I don't see the "alternatives" draft as a WG project,
I will try to work with Shawn and anyone else who is interested
to be sure that document contains a good description of the pros
and cons of just putting UTF-8 into the DNS rather than using an
ACE-style encoding.   Again, that is not a discussion we should
need to have anew each time people start thinking about IDNs.

Does that work for people?


More information about the Idna-update mailing list